Cyera, a data security company, has launched Cyera Research Labs, a dedicated research arm focused on the intersection of AI and data security, concurrently releasing its inaugural 2025 State of AI Data Security Report, which indicates that while 83% of enterprises use AI, only 13% have strong visibility into how AI interacts with sensitive data.
Cyera Research Labs aims to deliver clear, data-driven guidance on these emerging challenges. The 2025 State of AI Data Security Report was developed in partnership with CyberSecurity Insiders, surveying over 900 IT and security leaders to assess organizational readiness.
The report highlights several key findings regarding AI data security. Three-quarters of respondents, 76%, identify autonomous AI agents as the most challenging to secure, with only 9% monitoring AI activity in real time. Furthermore, only 16% of organizations treat AI as a distinct identity class with dedicated policies, while 21% grant broad data access to AI by default, contributing to potential vulnerabilities.
Control mechanisms appear to lag behind incidents; 66% of respondents reported having already detected AI over-accessing sensitive data. However, only 11% possess the capability to automatically block such risky activity. The study also points to immature governance frameworks, with just 7% of respondents having a dedicated AI governance committee and 11% feeling prepared for upcoming regulations. One participant recounted an instance where an AI copilot accessed a sensitive pricing deck due to default access and missing guardrails, underscoring the risks associated with inadequate monitoring at the prompt layer.
Shiran Bareli, VP of Research at Cyera, commented on these findings, stating, “These findings confirm what we’ve been hearing in the field from CISOs, CDOs, and other security professionals: AI adoption is outpacing governance, visibility, and controls. Too often, the discussion centers on the power of AI and the benefits it brings to the organization without equal focus on the data it consumes and exposes. With Cyera Research Labs, we want to bring data to the forefront of these conversations and highlight the pressing challenges security teams are grappling with every day.”
The 2025 State of AI Data Security Report establishes a cross-industry baseline for AI governance, aligning with the OWASP Top 10 for LLM Applications. It examines areas such as AI adoption, data visibility, monitoring practices, control implementation, access models, and overall governance readiness, offering security and data leaders practical guidance for implementing AI security programs. Cyera Research Labs states its mission is to analyze data creation, access, and sharing to uncover emerging threats and AI-driven risks, publishing insights that blend rigorous analysis with practical guidance.
Cyera describes itself as an AI-native data security platform designed to provide organizations with a comprehensive view of their data’s location, usage, and security status. The company aims to reduce risk and maximize data value across various environments. Cyera has secured over $1.3 billion in funding from investors including Accel, Coatue, Cyberstarts, Georgian, Lightspeed, and Sequoia. Its unified platform facilitates the discovery, securing, and leveraging of data, working to eliminate blind spots, reduce alert noise, and protect sensitive information across cloud, SaaS, databases, AI ecosystems, and on-premise environments. Recent platform enhancements, such as Cyera’s AI Guardian solution and Omni DLP, incorporate adaptive, AI-native data loss protection with real-time intelligence and contextual understanding of data movement within enterprises.
